After PV system installation, power consumption went drastically UP

[Goodi]

Hello,

My parents-in-law had a PV system with battery installed in September.
Before the PV system was installed, they had a power consumption of about 1500kWh to 2000kWh per year.
Afterwards they had a consumption of almost 800kWh in October alone. They are now already up to almost 2MWh since the installation.
And that's not just what the inverter is showing, they also got a horrendous bill from the power company.

I'm an EE engineer but my world is PCBs and high speed digital.

My mind struggles to understand how the system can appear to work properly and still cause this sudden increase in power consumption.
Any idea what could cause this?

Thank you very much in advance!
Greetings
Stefan

[coppice]

My guess is something has been reversed in measuring one of the current flows. Maybe a CT is clamped on a cable in a reversed manner, or two terminals have been swapped somewhere.

[nctnico]

There is zero export! Likely a problem in the metering

[Goodi]

As far as I know the system is not allowed to export power to the net. That's why.

[coppice]

There is zero export! Likely a problem in the metering

Why would they want to export anything, unless they have a massive surplus? There are batteries in the mix.

[Goodi]

But as I said also the meter of the power company shows increased consumption, thus the increased electrical bill. And as far as I know nobody touched anything there. The system is also not allowed to export power.

[coppice]

But as I said also the meter of the power company shows increased consumption, thus the increased electrical bill. And as far as I know nobody touched anything there. The system is also not allowed to export power.

Can you turn off the breaker from the PV panels for a couple of days, and see what that does to the numbers? If it doesn't restore the import to historic values it should tell you a lot.

[KE5FX]

Or walk around with a thermal cam.

[nctnico]

There is zero export! Likely a problem in the metering

Why would they want to export anything, unless they have a massive surplus? There are batteries in the mix.

But it is only a 5kWh battery by the looks of the daily charge capacity. That does nothing and it looks like this is a PV system somewhere between 5kWpeak to 10kWpeak which gives a massive surplus. My guess is the system is exporting (despite that it shouldn't do that) and the meter is adding imports and exports.

Another problem can be with the PV / battery inverter and the loads connected making the meter measuring wrong. Smart meters can be prone to that. Putting a good old (passive) meter with a rotary disk in series with the smart meter can be very useful to detect whether the meter is working correctly or not.

[rteodor]

If the meter is simple one without separate counters for import/export: follow the meter indication by hour and check if "consumption" increases while the sun is up compared to night time (or if the battery is empty). That should give a hint if there is an "unexpected export".

A clamp meter could be useful too to check what's coming from the grid vs. actual house consumption.

LE: But by the look of the daily variance, the "unintended export" is the most likely explanation.

[Siwastaja]

First make sure you have proper contract with grid/energy company for export; selling them your excess energy. You must have done that, right? Sure at that point someone also checked that your utility meter is capable of metering the export, right?

If that is taken care of, then make sure it works: get hourly metering data from the energy/grid company (sure it must be available online to you in year 2024 in Austria, yes?)

This must mean two numbers, export and import. Compare both to what the inverter application says, to rule out energy metering error.

Specifically check if there was any sunny day when you remember that battery was already showing 100% at some point in afternoon. Then there should be export. Verify that the export is recorded correctly both in the inverter and in the grid/energy company operator data.

Batteries are nasty in that they totally destroy your energy consumption if anything goes wrong with metering. In other words, the energy flow (direction and magnitude of power flow) the inverter sees MUST match what is measured by utility meter, pretty closely - especially close to zero.

And metering is not made easy by e.g. Chinese inverters having an overcomplicated system where a separate box is wired with RS485, then to that box is connected all three phase voltages and three CTs and the order of the phases must match as well as direction of CTs. Too many possibilities for installers to make a mistake, and mistake is hard to see, there is no intelligent diagnosis.

It is also important to know what is the netting period your grid company uses. This varies from milliseconds to a year depending on where you live and what kind of contract you have! And while of paramount importance, it can be nearly impossible to know what it is, you need insider information. For example, I know this is 15 minutes in most of Finland now, but I have no idea how I would Google that or where I would get that information as a consumer. Inverter net-zeroing feedback loop frequency response is somewhere around 0.1 - 0.5 Hz, so for netting periods shorter than a few minutes, extra cost is caused by timing mismatch. Also if phases are not netted together in a 3-phase system for billing (which was the case here still a few years ago!), it will totally ruin everything (inverter keeps total power at zero, which means some phases export while others import; if you are billed for the phases separately, you pay even at zero power).

Also see https://www.eevblog.com/forum/renewable-energy/low-power-efficiency-regulation-of-battery-hybrid-inverters-(eu-and-otherwise)/ - increased power consumption is to be expected.

I would suggest not to get grid-tied battery inverter systems without third party management system (aware of hourly energy prices; maybe participating in reserve markets). For PV timeshift only, it is not worth the investment, and extra power draw negates large part of the effect.

[Siwastaja]

There is zero export! Likely a problem in the metering

Why would they want to export anything, unless they have a massive surplus? There are batteries in the mix.

"Typical" 6-10kWp PV system of today produces some 30-40kWh a day, while typical battery size is 5 to 10 kWh. So most still goes to export. This is why for example our control algorithm often chooses to export even at very poor price (say 1 cent/kWh), if it is known from weather and consumption forecasts that battery would become full anyway, and forced to export at even worse price (say 0).

[brucehoult]

Before the PV system was installed, they had a power consumption of about 1500kWh to 2000kWh per year.

That seems incredibly low! That's somewhere between 170W and 230W, on average. My computers, fridge, and Starlink together use about that much -- not even counting lighting

Do they not use electricity for cooking or water heating? No heat pump, no heaters?

Afterwards they had a consumption of almost 800kWh in October alone.

That's about 30% more than the average house here in NZ.

I don't know what kind of place they live in, and you didn't provide any details, but I would find that more believable than 2000 kWh/year, unless they're living in a small apartment with gas cooking and hot water and heating provided by the apartment complex (and charged for separately).

But if it's a hi-rise apartment complex then where didn't they put 9 kWh a day (in October!) worth of PV panels?

[Zero999]

Before the PV system was installed, they had a power consumption of about 1500kWh to 2000kWh per year.

That seems incredibly low! That's somewhere between 170W and 230W, on average. My computers, fridge, and Starlink together use about that much -- not even counting lighting

Do they not use electricity for cooking or water heating? No heat pump, no heaters?

I dissagree. I wouldn't say it's incredibly low. Those figures are double mine.

My annual energy usage: July 2021 to 2022


I live in a small two bedroom property. I have gas heating and a gas stove. I also use a microwave, but it's not much.

Gas is more commonly used for heating and cooking in Europe. Heat pumps are being pushed to cut pollution, but are still not that common.

[Siwastaja]

But the consumption is incredibly low to buy PV let alone a battery system. That's the mistake here. These systems usually make sense in the 10MWh/year range and above. With so little consumption, PV goes mostly to export, so it only works in countries where special incentives for exporting are available; in most places this is not the case.

And with so little consumption, battery would be mostly used to buy and sell energy on spot price markets, or act in frequency reserve markets. Which is something a "normal" hybrid inverter + battery simply does not do at all; there are specialized companies and products doing that.

[nctnico]

Batteries are sold by many scammers nowadays. I came across this video which warns against these scams yesterday:



It is an item from a TV show which deals with all kinds of consumer related items and scams. It is in Dutch; no idea whether Youtube can add subtitles in different languages or that it is even available outside the NL.

People are pressured to buy a battery for nearly twice the price and borrow money through a 'zero interest' loan. They also need to buy the electricity from the same company in order to get to the promised ROI. The small print in the contract however says the complete opposite of the sales pitch.

[Siwastaja]

Batteries are sold by many scammers nowadays.

We have interesting situation here as well; not something I would (at least publicly) call scammers, but let's say borderline scammers. These are supposedly "well known" "large" companies who push their stuff to random small "three electricians and a van" companies to sell it to the customers, especially the elderly (who also have money). Here the big buzzword at the time is frequency reserve markets. Which is a real market of course, but the size of it is 1/100th of e.g. spot price market and is rather quickly saturated with batteries.

The market collapsed in Sweden a few years ago; initially, large paychecks (like 200 EUR/month) were promised and some actually got those for some months but then it melted to a level of maybe 10-20 EUR/month; the company doing it lost its reputation quite badly and the whole industry got a hit.

We are doing battery control now but to make it non-insane one needs to combine optimization of own PV storage to using hourly spot energy markets and optimize the battery usage carefully as a whole. Even then the investment is large and payback is long. I used to say it never makes any sense to invest in batteries but after doing simulations of our optimization algorithm I can say that it's not completely stupid. And participating into reserve markets can add some earnings; if they dilute, no problem, because energy price variations during the day are not really going anywhere.

You may want to take a look of my post about this with your preferred translation tool: https://www.enion.fi/artikkelit/enion-ohjaa-akkuja-lykksti-jo-tnn

[nctnico]

IMHO it greatly depends on where you are located whether large price variations during the day will exist for a long time or not. Over here in the NL various companies are building large storage facilities using batteries and you can't beat their economy of scale as a tiny consumer. Especially not with a company in the middle as a broker / service provider getting part of the profit (which is probably similar to your solution; a paid for service to determine the best time to charge & discharge the battery).  So I expect extreme price variations to be a thing of the past in a few years in the NL; at least it is not something I'd recommend basing in investment on with a projected ROI >3 years. Remember: the people getting very rich from mining are the ones selling the mining equipment. 

[Siwastaja]

It is indeed impossible to guarantee payback for a battery system. Though, if one is getting a battery only for PV timeshift  - I can pretty much guarantee the impossibility of any payback. Being such expensive investment, one needs to utilize the battery in every possible way instead of letting it sit with light use.

There are some benefits of distributing the batteries at consumer sites: while centralized sites require stronger grid infrastructure, consumer batteries have potential to do the opposite (reduce peak load) - although peak charge rate when everybody charges at the single cheapest hour might then again nullify this advantage (demonstrating the complexity of optimizing everything). This also shows in money: if you are a consumer and use your own produced PV, you get more out of it compared to a centralized plant pushing electricity into grid. Details depend on country but for example here the bill is separated into energy and grid fee; by exporting you don't get paid for the grid fee; import price is bigger than export price. So with centralized solution grid is in use, and the consumer will pay for it, in one way or another. Then there is also emotional side; people feel they are more self-sufficient, or that they are participating in the energy (r)evolution.

But you are probably right that even with the mentioned benefits it is more cost-effective to do large-scale storage. And not only for getting the batteries and inverters cheaper per kWh, but also because large scale pretty much forces you to do things right. Home installations sit at poor utilization, or even worse, doing damage to the customer (e.g. by miswired/misconfigured energy meter). In my opinion grid-tied battery systems should not be sold at all without intelligent control because even with the control it's an iffy market.